This research roundup gathers the ten most important studies on depression treatment, from the landmark STARD trial through the 2025 Dartmouth generative AI therapy trial published in NEJM AI. Each entry names the researchers, explains the study design, summarizes the findings, and provides citations. Depression treatment research has evolved substantially over the past two decades, moving from a strong assumption that SSRIs were first-line for nearly everyone to a more nuanced picture in which cognitive behavioral therapy, behavioral activation, exercise, mindfulness-based cognitive therapy, and now AI-delivered interventions each have solid evidence. The studies below include the STARD trial that showed remission with a single antidepressant is not the norm, Pim Cuijpers's 2023 meta-analysis that updated psychotherapy effect sizes, the NEJM trial showing exercise matched SSRIs, the Dartmouth AI therapy trial, Alison Darcy's Woebot RCT, the CBT meta-analyses, TMS studies, ketamine research from Yale, behavioral activation, and Zindel Segal's mindfulness-based cognitive therapy for relapse prevention. If you or someone you love has depression, knowing which treatments have the strongest evidence helps you ask the right questions of a provider. Citations include original venues. The convergent message: depression is treatable, multiple approaches work, and combining approaches is often more effective than any single one. The 2025 Dartmouth and 2024 Cuijpers findings have begun to reshape the field.

1. What Did the STAR*D Trial Find?

The STAR*D trial (Sequenced Treatment Alternatives to Relieve Depression), led by John Rush and colleagues and published in multiple papers from 2006 onward in the American Journal of Psychiatry, followed nearly 4 thousand depressed outpatients through a series of treatment steps. It found only about one third achieved remission with the first antidepressant (citalopram). Cumulative remission across four steps was about 67 percent. It matters because it revealed how often depression is treatment-resistant. Citation: Rush et al., American Journal of Psychiatry (2006).

2. What Did Cuijpers 2023 Meta-Analysis Conclude?

Pim Cuijpers at VU Amsterdam led a 2023 network meta-analysis of psychotherapy for adult depression published in World Psychiatry. Analyzing hundreds of trials, it found that cognitive behavioral therapy, interpersonal therapy, behavioral activation, and problem-solving therapy all showed comparable efficacy, with effect sizes around 0.7 to 0.9 on the standardized mean difference. It matters because it confirmed psychotherapy is at least as effective as medication for most patients. Citation: Cuijpers et al., World Psychiatry (2023).

3. What Did the NEJM Exercise vs SSRI Trial Find?

A 2023 and 2024 wave of studies synthesizing exercise research for depression found that structured exercise programs produced effect sizes comparable to SSRIs and cognitive behavioral therapy for mild to moderate depression. The BMJ meta-analysis by Noetel and colleagues (2024) covered 218 studies and confirmed exercise as a first-line treatment. It matters because it established exercise as an evidence-based treatment, not a nice-to-have add-on. Citation: Noetel et al., BMJ (2024).

4. What Did the Dartmouth Therabot NEJM AI Trial Show?

Michael Heinz, Nicholas Jacobson and colleagues at Dartmouth Geisel School of Medicine published in NEJM AI (2025) the first large randomized controlled trial of a generative AI chatbot (Therabot) for depression, anxiety, and eating disorder symptoms. Participants using Therabot showed statistically significant symptom reductions compared to a waitlist control. It matters because it provided the first NEJM-level evidence for generative AI therapy. Citation: Heinz et al., NEJM AI (2025).

5. What Did the Woebot 2017 RCT Establish?

Alison Darcy, Kathleen Fitzpatrick and colleagues at Stanford and Woebot Health ran a two-week randomized controlled trial of Woebot, a CBT-informed chatbot, published in JMIR Mental Health in 2017. College students using Woebot showed significant reductions in PHQ-9 depression scores compared to information control. It was the founding clinical evidence for therapeutic chatbots. Citation: Fitzpatrick, Darcy and Vierhile, JMIR Mental Health (2017).

6. What Do CBT Meta-Analyses Show?

Cognitive behavioral therapy has been studied more than any other psychological treatment for depression, with dozens of meta-analyses over three decades. Major syntheses by Cuijpers and Hofmann have consistently shown moderate to large effect sizes for CBT versus waitlist and comparable to medication over the long term, with better relapse prevention. It matters because CBT is the benchmark against which all other depression treatments are measured. Citation: Hofmann et al., Cognitive Therapy and Research (2012).

7. What Did TMS Studies Show?

Transcranial Magnetic Stimulation (TMS) has been extensively studied for treatment-resistant depression since its FDA approval in 2008. Meta-analyses (Berlim et al. 2014 in Psychological Medicine and others) showed response rates around 29 percent and remission rates around 18 percent in treatment-resistant patients. SAINT protocol studies from Stanford (Cole et al. 2020) showed higher rates with accelerated protocols. It matters because it offers hope for patients who have failed medication. Citation: Cole et al., American Journal of Psychiatry (2020).

8. What Did Yale Ketamine Research Contribute?

Researchers at Yale led by John Krystal, Dennis Charney, and Gerard Sanacora showed in the early 2000s that a single IV ketamine infusion produced rapid antidepressant effects in treatment-resistant depression, often within hours. Subsequent studies confirmed the effect and led to the FDA approval of esketamine nasal spray in 2019. It matters because it opened an entirely new class of rapid-acting antidepressants. Citation: Berman et al., Biological Psychiatry (2000) and Zarate et al., Archives of General Psychiatry (2006).

9. What Did Behavioral Activation Research Establish?

Behavioral activation, developed by Neil Jacobson and refined by Christopher Martell, focuses on increasing engagement in rewarding activities rather than changing thoughts. Jacobson's dismantling study (1996) showed BA was as effective as full CBT. A 2016 large UK trial (Richards et al., Lancet) confirmed BA matched CBT but was cheaper and easier to deliver. It matters because it provides an accessible evidence-based treatment. Citation: Richards et al., The Lancet (2016).

10. What Did Mindfulness-Based Cognitive Therapy Show for Relapse Prevention?

Zindel Segal, Mark Williams and John Teasdale developed Mindfulness-Based Cognitive Therapy (MBCT) in the late 1990s to prevent depression relapse. Their 2002 book and subsequent randomized trials showed MBCT reduced relapse rates by about 40 percent in patients with three or more prior episodes. A 2016 meta-analysis (Kuyken et al., JAMA Psychiatry) confirmed efficacy. It matters because it addressed the problem of chronic recurrent depression. Citation: Kuyken et al., JAMA Psychiatry (2016). These ten studies represent the evidence base for modern depression treatment. Taken together, they suggest depression is genuinely treatable, multiple evidence-based options exist, and treatment resistance does not mean hopelessness because later steps including TMS, ketamine, and newer chatbot approaches can work when earlier steps fail. If you or someone you love is facing depression, knowing these studies helps you ask the right questions of a provider about options and evidence.